Mathew Hall

About

Mathew Hall is a leading researcher at the intersection of computer vision and reconfigurable hardware, with a primary focus on real-time object detection and FPGA-based acceleration. His most influential work, "End-to-End FPGA-based Object Detection Using Pipelined CNN and Non-Maximum Suppression" (2021, 16 citations), addresses a critical challenge in deploying deep learning for edge applications: achieving high-speed, low-latency inference on resource-constrained devices. Hall’s major contribution lies in designing a fully pipelined architecture that integrates a convolutional neural network (CNN) for feature extraction with a single-shot detector (SSD) and non-maximum suppression directly on FPGA fabric. This end-to-end approach eliminates the bottleneck of off-chip processing, enabling efficient detection, classification, and localization for autonomous driving, smart surveillance, and robotics. By demonstrating that complex computer vision pipelines can be realized on reconfigurable logic without sacrificing accuracy, Hall’s work has paved the way for practical, real-time embedded vision systems. His research continues to bridge the gap between algorithmic advances in deep learning and the hardware constraints of edge devices, making him a key figure in the growing field of FPGA-accelerated AI.

Research Focus

Key Achievements

Top Papers

1. 1
   End-to-End FPGA-based Object Detection Using Pipelined CNN and Non-Maximum Suppression

   16 citations · 2021

Key Collaborators

Contact & Links